Every story and comment from the Hacker News API, in a layout borrowed from 2001.
A comment on Qwen 3.6 27B is the sweet spot for local development. Parent comment
With dflash-mlx library, with dflash disabled 3.6-35b-a3b mlx model I'm getting ~60 t/s on low power and >100 t/s on high power.
Compared with previous message being official qwen huggingface release (non-mlx) using lm studio.
There are no comments on this comment yet.
With dflash-mlx library, with dflash disabled 3.6-35b-a3b mlx model I'm getting ~60 t/s on low power and >100 t/s on high power.
Compared with previous message being official qwen huggingface release (non-mlx) using lm studio.